GKE Autopilot

# GKE Autopilot 클러스터 생성
gcloud container clusters create-auto my-autopilot \
  --region=asia-northeast3 \
  --release-channel=regular \
  --network=default \
  --subnetwork=default

# 클러스터 인증 정보 가져오기
gcloud container clusters get-credentials my-autopilot \
  --region=asia-northeast3

# Autopilot 클러스터 정보 확인
gcloud container clusters describe my-autopilot \
  --region=asia-northeast3 \
  --format="value(autopilot.enabled)"

# Private Autopilot 클러스터 생성
gcloud container clusters create-auto my-private-autopilot \
  --region=asia-northeast3 \
  --release-channel=regular \
  --enable-private-nodes \
  --enable-master-authorized-networks \
  --master-authorized-networks=10.0.0.0/8

# GKE Autopilot 클러스터 Terraform 구성
resource "google_container_cluster" "autopilot" {
  name     = "my-autopilot-cluster"
  location = "asia-northeast3"

  # Autopilot 모드 활성화
  enable_autopilot = true

  # 릴리스 채널 설정
  release_channel {
    channel = "REGULAR"
  }

  # VPC 네이티브 설정 (Autopilot 필수)
  ip_allocation_policy {}

  # Private 클러스터 설정
  private_cluster_config {
    enable_private_nodes    = true
    enable_private_endpoint = false
    master_ipv4_cidr_block  = "172.16.0.0/28"
  }

  # 승인된 네트워크
  master_authorized_networks_config {
    cidr_blocks {
      cidr_block   = "10.0.0.0/8"
      display_name = "internal"
    }
  }
}

# 출력
output "cluster_endpoint" {
  value = google_container_cluster.autopilot.endpoint
}

# Autopilot에 적합한 Deployment 예제
# 리소스 요청이 반드시 필요 (과금 기준)
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-service
spec:
  replicas: 3
  selector:
    matchLabels:
      app: web-service
  template:
    metadata:
      labels:
        app: web-service
    spec:
      containers:
      - name: web
        image: gcr.io/my-project/web:v1
        ports:
        - containerPort: 8080
        resources:
          requests:
            cpu: "250m"      # Autopilot 과금 기준
            memory: "512Mi"  # 최소 512Mi 권장
          limits:
            cpu: "500m"
            memory: "1Gi"
        # Autopilot 호환 프로브
        readinessProbe:
          httpGet:
            path: /health
            port: 8080
          initialDelaySeconds: 10
          periodSeconds: 5
---
# Autopilot에서 GPU 워크로드
apiVersion: v1
kind: Pod
metadata:
  name: gpu-workload
spec:
  nodeSelector:
    cloud.google.com/gke-accelerator: nvidia-tesla-t4
  containers:
  - name: gpu-container
    image: nvidia/cuda:11.0-base
    resources:
      limits:
        nvidia.com/gpu: 1